Zaphs' latest thing on non-domes

[aaronm]

Have you seen Zaphs' tests on
the planars and ribbons? The BG neo won by a landslide
(I guess I'll be selling my Arum Cantus soon, Shhh!
don't tell anyone they suck!)
I REALLY want to know the lowdown on the BG Neo8PDR,
if they're half as good as the Neo3, I just might use
those for my line/dipole...any thoughts?

[dst]

> Have you seen Zaphs' tests on
> the planars and ribbons? The BG neo won by a
> landslide
> (I guess I'll be selling my Arum Cantus
> soon, Shhh!
> don't tell anyone they suck!)
> I REALLY want to know the lowdown on the BG
> Neo8PDR,
> if they're half as good as the Neo3, I just
> might use
> those for my line/dipole...any thoughts?

NEO 8 does not have the extension of the NEO 3. Rolls off early and has a bump at around 12K that needs attenuation. Many consider it a midrange driver vs. a true tweeter.

[aaronm]

> NEO 8 does not have the extension of the NEO
> 3. Rolls off early and has a bump at around
> 12K that needs attenuation. Many consider it
> a midrange driver vs. a true tweeter.

After the peak is dealt with, do you know how high the Neo8 really goes? It looks like it behaves very similar to the RD series drivers.

Something to consider, I have proved to myself that my ears not register ANYthing above 13k.
That said, I don't think I need a ton of HF extension.

I'd like to use six per side as tweets next to a line of a dozen Focal 5V4411s (per side)

[raw]

> NEO 8 does not have the extension of the NEO
> 3. Rolls off early and has a bump at around
> 12K that needs attenuation. Many consider it
> a midrange driver vs. a true tweeter.

Ya kind of hard to do distortion tests on a Neo that only goes up to 18k on axis to a ribbon that measures to 27k?
As well as distortion measured at 2k lets get real!
You want to test for tweeter distortion test them all above a normal crossover point!Were a tweeter is actually rolled in after a crossover network.

Let ask John to do thos test again.
But do the following!
Put them in a crossover all at 3000-4000 as this is the range that a majority of the public would use a crossover point.
Measure all the drivers with a crossover at 7k,10k and 15k were the distortion counts on a tweeter not at 2k.If you want to compare 2k distortion do so on midgrange speakers not tweeters!

And last do off axis measurments 0-45 and then lets see what comes out on top!
You want to know before he does the tests!
Try the ribbons will walk all over that Neo for off axis measurements.
As well as distortion in the upper spectrum as well.

Comes down to this. If you are doing distortion tests on a tweeter and want to compare tweeters compare the frequency they are used not at 2k.
All tweeters run out of suspention around 2k or lower.

Do a real tweeter test.

[chichi]

I'm guessing that you have made measurements using such criteria. Please post them if so. I have the worst one of the bunch, but with a 3k 12dB/octave crossover, it sounds alright to me.

[Pete Schumacher ®]

> Ya kind of hard to do distortion tests on a
> Neo that only goes up to 18k on axis to a
> ribbon that measures to 27k?
> As well as distortion measured at 2k lets
> get real!
> You want to test for tweeter distortion test
> them all above a normal crossover point!Were
> a tweeter is actually rolled in after a
> crossover network.

> Let ask John to do thos test again.
> But do the following!
> Put them in a crossover all at 3000-4000 as
> this is the range that a majority of the
> public would use a crossover point.
> Measure all the drivers with a crossover at
> 7k,10k and 15k were the distortion counts on
> a tweeter not at 2k.If you want to compare
> 2k distortion do so on midgrange speakers
> not tweeters!

> And last do off axis measurments 0-45 and
> then lets see what comes out on top!
> You want to know before he does the tests!
> Try the ribbons will walk all over that Neo
> for off axis measurements.
> As well as distortion in the upper spectrum
> as well.

> Comes down to this. If you are doing
> distortion tests on a tweeter and want to
> compare tweeters compare the frequency they
> are used not at 2k.
> All tweeters run out of suspention around 2k
> or lower.

> Do a real tweeter test.

Using a crossover will have NO IMPACT on distortion measurements with pure tones. Instead of being so indignant about the performance of the ribbons under these test conditions, just accept the fact that most of those ribbons should not be used in a TM two-way. They just don't perform well when asked to go below 3000-4000Hz.

Where they should be used is in line arrays.

I hope you understand why adding a crossover won't help in determining distortion levels. If you need me to explain that further, I'd be happy to elaborate.

Measuring distortion at 10KHz is pointless as the first harmonic above the fundamental is beyond human hearing. Distortion at 3KHz is very important as there are harmonics at 6, 9, 12, 15, and 18KHz.

And before you say that the ribbons will blow away the Neo off axis, you may want to measure yourself. An element as small as the Neo will certainly have quite good off-axis response, and certainly it's vertical off-axis will be far better than the other ribbons.

[raw]

> Using a crossover will have NO IMPACT on
> distortion measurements with pure tones.
> Instead of being so indignant about the
> performance of the ribbons under these test
> conditions, just accept the fact that most
> of those ribbons should not be used in a TM
> two-way. They just don't perform well when
> asked to go below 3000-4000Hz.

> Where they should be used is in line arrays.

> I hope you understand why adding a crossover
> won't help in determining distortion levels.
> If you need me to explain that further, I'd
> be happy to elaborate.

> Measuring distortion at 10KHz is pointless
> as the first harmonic above the fundamental
> is beyond human hearing. Distortion at 3KHz
> is very important as there are harmonics at
> 6, 9, 12, 15, and 18KHz.

> And before you say that the ribbons will
> blow away the Neo off axis, you may want to
> measure yourself. An element as small as the
> Neo will certainly have quite good off-axis
> response, and certainly it's vertical
> off-axis will be far better than the other
> ribbons.

Measuring distortion at 10KHz is pointless as the first harmonic above the fundamental is beyond human hearing. Distortion at 3KHz is very important as there are harmonics at 6, 9, 12, 15, and 18KHz.

My point if you are measuring tweeter distortion to show which is better do so over a range just like you also just said.
6,9,12,15,18khz.

Doing tests at a single freqency shows nothing of the over all performance of a product.
Also as I said all data should be posted including 0-45 off axis!

Nothing against the tests, but jumping up and down to say this driver over that one is not a fact till you show all the facts not 1% of them.

As for the 3k point I still stand yes this can be important for some but my point is why!
That is in the middle or is the crossover point for the drivers all tested!

And last comparing the long ribbon driver to a short is like comparing a 4" to a 12" driver.

IF you are to compare the ribbons do so on the length of the ribbon.
We will say that for this test should have done the G2 and G2Si as these are still longer that the 1" tweeters tests but close enough to say so.
The shorter LCY would also be included.

Then do a test comparing longer ribbons.

[Jim Holtz]

A few months ago Mark K. tested Neo3's and they tested poorly enough that he chose not to post the results if memory serves me correctly. I'm curious why John and Marks results aren't the same.

I'm not going to get into a p*ss*ng match over measurements but audibility and dispersion should be the rest of the discussion. There are a whole lot of folks whose ears I trust that think ribbons sound extremely nice and *ACCURATE* to the original performance. I happen to agree with them.

Start with measurements, end with your ears. :-)

Jim

> Using a crossover will have NO IMPACT on
> distortion measurements with pure tones.
> Instead of being so indignant about the
> performance of the ribbons under these test
> conditions, just accept the fact that most
> of those ribbons should not be used in a TM
> two-way. They just don't perform well when
> asked to go below 3000-4000Hz.

> Where they should be used is in line arrays.

> I hope you understand why adding a crossover
> won't help in determining distortion levels.
> If you need me to explain that further, I'd
> be happy to elaborate.

> Measuring distortion at 10KHz is pointless
> as the first harmonic above the fundamental
> is beyond human hearing. Distortion at 3KHz
> is very important as there are harmonics at
> 6, 9, 12, 15, and 18KHz.

> And before you say that the ribbons will
> blow away the Neo off axis, you may want to
> measure yourself. An element as small as the
> Neo will certainly have quite good off-axis
> response, and certainly it's vertical
> off-axis will be far better than the other
> ribbons.

[dlr]

> Using a crossover will have NO IMPACT on
> distortion measurements with pure tones.

Yes it will, especially of the linear variety. You might ask John K why he has been so adament about the distortion tests not being useful without a crossover.

I don't care one way or the other which one may be deemed better or not, but distortion can be impacted by the crossover.

In fact, I was debating John about the impact of response anomolies that are multiples of a fundamental (pure tone) and that it keeping the Fc of a lowpass two octaves or more away as the crossover cannot control the harmonic distortion amplification, but he emphatically stated that if the crossover lowpass matched the target that it the harmonic distortion was controlled.

The same would hold for a highpass, if some non-linearities in the response were corrected, then any distortion where the multiple coincided with the peak would also be controlled.

Now non-linear distortion, that's a separate issue.

dlr

[Jeff B.]

> Yes it will, especially of the linear
> variety. You might ask John K why he has
> been so adament about the distortion tests
> not being useful without a crossover.

> I don't care one way or the other which one
> may be deemed better or not, but distortion
> can be impacted by the crossover.

> In fact, I was debating John about the
> impact of response anomolies that are
> multiples of a fundamental (pure tone) and
> that it keeping the Fc of a lowpass two
> octaves or more away as the crossover cannot
> control the harmonic distortion
> amplification, but he emphatically stated
> that if the crossover lowpass matched the
> target that it the harmonic distortion was
> controlled.

> The same would hold for a highpass, if some
> non-linearities in the response were
> corrected, then any distortion where the
> multiple coincided with the peak would also
> be controlled.

> Now non-linear distortion, that's a separate
> issue.

> dlr

I have been following these threads at Mad and here, but so far have kept myself out of them. I have an aweful lot I would like to say, however, and some of it is along the same lines of what you and John have stated. The rest would be primarily from my own experience and would run counter to most of what Pete has saying on this topic. The perspective here has been way too narrow in my opinion.

[rickcraig]

Provided Link: Selah Audio


> I have been following these threads at Mad
> and here, but so far have kept myself out of
> them. I have an aweful lot I would like to
> say, however, and some of it is along the
> same lines of what you and John have stated.
> The rest would be primarily from my own
> experience and would run counter to most of
> what Pete has saying on this topic. The
> perspective here has been way too narrow in
> my opinion.

Sounds like something good to discuss when we all meet in Indianapolis